ESTRO 35 Abstract-book

S440 ESTRO 35 2016 ______________________________________________________________________________________________________

Comparing with the standard IMRT scenario, the most effective ART scenario was the one with six replannings, leading to a decrease of 4.94 Gy (12.22Gy max.) 86 % of this benefit was obtained with 3 replannings only (at week 1-2-5). If only one replanning should be applied, it should be done at the first week. Conclusion: Each supplementary replanning leads to a decrease of the mean PG dose. Early replanning proved the most beneficial for sparing the PG. Considering the maximum benefit obtained with six replannings, almost 90% of this benefit was obtained with only three replannings (Weeks 1-2-5), thus representing an attractive combination for ART in locally-advanced HNC. Poster: Physics track: CT Imaging for treatment preparation PO-0912 MRI-only based RT: adopting HU conversion technique for pseudo-CT construction in various body parts J. Korhonen 1 Helsinki University Central Hospital, HUCH Cancer Center, Helsinki, Finland 1 , L. Koivula 1 , T. Seppälä 1 , M. Kapanen 1 , M. Tenhunen 1 Purpose or Objective: MRI is increasingly applied for radiotherapy target delineation. Recent studies have demonstrated a possibility to omit CT imaging from the radiotherapy treatment planning workflow by developing methods enabling the entire process by relying on MRI only. The HU conversion technique has been shown to construct heterogeneous CT-representative (pseudo-CT) images for prostate cancer patients by transforming the intensity values of an in-phase MR image into HUs with separate conversion models for soft and bony tissues. The technique has been implemented into a routine MRI-only based radiotherapy treatment planning workflow in our clinic. This study aims to investigate whether the pseudo-CT construction technique could be adopted for different patient groups, also in different body sites in addition to the male pelvis. Material and Methods: The examinations were conducted by investigating the correspondence between the MR image intensity values and CT image HUs for different tissues. The data were applied to develop HU conversion models to transform the MR image intensities into appropriate HUs. In the absence of air cavities, the method was applied as a dual model HU conversion technique with separate conversion models within and outside of a bone segment obtained by atlas and threshold -based segmentation methods. An additional air segment was constructed in the presence of air cavities. An ultra-short echo-time sequence was applied to recognize boundaries between air and bone cortex with an intensity threshold. The constructed HU conversion models were employed with a medical image processing software, and applied for head (10 patients), pelvis (10), abdomen (2), and limbs (2). The obtained pseudo-CT images were tested by comparisons against standard CT images. The tests included evaluation of HU uncertainty and photon dose calculation accuracy. Results: The HU conversion technique enabled construction of heterogeneous pseudo-CT images for various body sites. The duration of MR image intensity value transformation into HUs was roughly 30 seconds for each image series. Figure 1 shows examples of the resulted pseudo-CT images with the original MR images. Table 1 presents the local HU differences between those in pseudo-CT images and those in standard CT images. The target volume mean dose differences between those in pseudo-CT images and those in standard CT images were within 1% in all cases.

Conclusion: The HU conversion technique can be adopted for various body sites to enable construction of heterogeneous pseudo-CT images for MRI-only based radiotherapy treatment planning. The conversion models should be adjusted for each site separately to improve pseudo-CT image quality; e.g. for abdomen. Further examinations are ongoing. PO-0913 Clinically applicable T2-weighted 4D Magnetic Resonance Imaging with good abdominal contrast D. Tekelenburg 1 Academic Medical Center, Department of Radiation Oncology, Amsterdam, The Netherlands 1 , O. Gurney-Champion 1 , E. Lens 1 , A. Van der Horst 1 , A. Nederveen 2 , A.K. Biegun 3 , A. Bel 1 , Z. Van Kesteren 1 2 Academic Medical Center, Department of Radiology, Amsterdam, The Netherlands 3 University of Groningen, KVI-Center for Advanced Radiation Technology, Groningen, The Netherlands Purpose or Objective: The main drawback of CT is poor soft tissue contrast. This research aims to develop an accurate respiratory-correlated four-dimensional MRI (4D MRI) method analogous to 4D CT with a clinically relevant acquisition time and superior contrast for abdominal structures. Material and Methods: We developed a 4D MRI method by alternating a fast (0.6 seconds per 2D slice) T2-weighted turbo spin echo image acquisition (resolution: 1.3 x 1.6 mm²; 5 mm thickness) with a 1D navigator acquisition. The navigator obtained the diaphragm position prior to each slice acquisition. The total acquisition was done continuously during free breathing for 6 minutes, covering multiple respiratory cycles and yielding 60 image frames per slice over a volume of 11 slices.